This research continues the exploratory work recently completed on the use of diaries for collecting health data in household surveys. Current recall procedures suffer from memory errors, omissions and telescoping, that diaries eliminate. Diaries will be tested on a general population of Illinois and compared to recall methods. Four strata each with 300 households will be selected after an initial screening: high level of medical experience in the previous year, low education; high experience, high education; low experience, low education; low experience, high education. The three treatments will be: three personal interviews at monthly intervals, recruit to keep a diary for three months with compensation, recruit to keep a diary without compensation. Phone interviews and mail-back diaries will also be tested. The analysis will compare cooperation, levels of reporting and costs for the alternative strata and treatments.